Diet-induced impaired glucose tolerance and gestational diabetes in the dog

Abstract

Glucose metabolism was compared in dogs consuming a chow/meat diet throughout pregnancy (P group, n = 6) and dogs switched to a high-fat/high-fructose (HFF) diet during the 4th-5th gestational week (gestation ≃9 wk; P-HFF group; n = 6). An oral glucose tolerance test (OGTT; 0.9 g/kg) was administered in the 6th-7th gestational week, and a hyperinsulinemic [0-120 min: 1.8 pmol·kg(-1)·min(-1) (low insulin); 120-240 min: 9 pmol·kg(-1)·min(-1) (high insulin)] euglycemic clamp was performed the following week. Nonpregnant (NP) female dogs underwent OGTTs but not clamp studies. All P-HFF dogs exhibited impaired glucose tolerance (IGT) or gestational diabetes (GDM), but only one P dog had IGT. Insulin concentrations in P and P-HFF dogs were significantly lower than in NP dogs 30 and 60 min after the OGTT. Therefore, mean islet size and area were evaluated in P and NP dogs. These values did not differ between groups, and proliferating endocrine cells were rare in pregnancy. During exposure to high insulin, glucose infusion rate and hindlimb glucose uptake were ∼30% greater (P < 0.05) and net hepatic glucose output was more suppressed (-5.5 ± 6.1 vs. 7.8 ± 2.8 mg·100 g liver(-1)·min(-1), P < 0.05) in P than in P-HFF dogs. In conclusion, in the 2nd trimester the canine pancreas does not exhibit islet hypertrophy, hyperplasia, or neogenesis. Combined with the lack of pancreatic adaptation, a HFF diet during late pregnancy produces a canine model of IGT and GDM without hyperinsulinemia but exhibiting liver and muscle insulin resistance.

Fig. 1.

Deep venous plasma concentrations of glucose (A) and insulin (B) in nonpregnant (NP; n = 3, shaded circles and lines), normal pregnant (P; n = 4, open squares and solid line), and high fat/high fructose-fed overnight-fasted conscious dogs [P-HFF; a combined group (n = 6, filled triangles and dashed lines) including 4 dogs with normal pancreata and 2 with partial pancreatectomy, as well as a P-HFF subset (n = 4, open triangles and dashed-dotted lines) containing only the dogs with normal pancreata] in response to an oral glucose load (0.9 g/kg).

Fig. 2.

Euglycemic hyperinsulinemic clamp data in P group (n = 6, open bars), P-HFF combined group (n = 6, filled bars), and P-HFF subset with normal pancreata (n = 4, hatched bars): arterial plasma glucose concentrations (A), net hepatic glucose balance (B), hindlimb glucose uptake (C), nonhepatic glucose uptake (D), and glucose infusion rate (E). *P < 0.05 vs. P group.

Fig. 3.

Total islet size distribution (A), average islet size (B), and total islet area (C) in normal pregnant (late 2nd trimester; striped bars) and nonpregnant (open bars) female dogs (n = 3 per group). There were no significant differences between groups.